Vidlora guide

How to extract audio from a video

Extracting audio separates the sound from the picture and saves it as its own file. It is useful for podcasts recorded as video, lectures you want to listen to on a commute, interview transcripts, and soundtracks for your own projects. The details matter: format, bitrate and sample rate decide how the result sounds and how large it is. This guide covers each choice with numbers, explains when extraction is lossless, and lists the mistakes to avoid.

7 minute read

What extraction actually does

A video file is a container with at least two streams: video and audio. Extraction reads the audio stream and writes it to a new file, ignoring the video. Depending on settings, it either copies the stream unchanged, which is fast and lossless but limited to the format the source already uses, or decodes it and re-encodes it as another format such as MP3. Vidlora's extraction tool discards the video and encodes the audio to the output format you choose, so the result contains sound only. Because the picture is gone, the file shrinks dramatically: a 500 MB, 10-minute video may yield a 10 MB audio file, since audio at 128 kbps uses about 1 MB per minute.

Why not just mute or screen-record

Muting a video keeps the picture and its large file size; it is for removing sound, not extracting it. Screen-recording the sound while a video plays records through your device's audio chain, picks up system notifications and gains an extra generation of lossy compression, and it only works in real time. Extraction is better on every count: it is faster than playback, free of interruptions, and starts from the original audio stream. It also preserves the timing, so the audio file will line up exactly with the video if you need to reattach it later. For a one-hour recording, extraction typically completes in well under a minute, while screen-recording needs the full hour.

MP3, AAC and WAV compared

MP3 is the compatible default: it plays on every device and suits speech and general music. AAC, usually saved as .m4a, gives somewhat better quality at the same bitrate and is the native format for iPhones and many video platforms; if the source audio is already AAC, extraction to AAC avoids an extra lossy step. WAV is uncompressed PCM: at 44.1 kHz and 16-bit stereo it uses roughly 10.6 MB per minute, so an hour is over 600 MB, but it is the right choice if you will edit, mix or run speech recognition on the audio. Choose MP3 for sharing and listening, AAC for Apple devices or efficiency, and WAV for editing. Avoid converting between lossy formats repeatedly.

Choosing a bitrate

For speech, such as podcasts, lectures, meetings and interviews, 96 to 128 kbps is clear and keeps files small: one hour at 96 kbps is about 43 MB, at 128 kbps about 58 MB. For music, use 192 kbps as a good default and 256 to 320 kbps if the source is high quality and you have space; 320 kbps costs about 144 MB per hour. Do not choose a bitrate higher than the source offers, because it only inflates the file. If you are not sure, check the original audio bitrate and codec in a metadata viewer; if it reports around 128 kbps AAC, a 128 to 192 kbps output is sufficient. For voice notes where size is critical, 64 kbps mono is workable.

Sample rate and channels

Sample rate is how many measurements per second describe the sound. 44.1 kHz is the CD standard and 48 kHz is the standard for video; both capture frequencies up to what most people can hear. Speech is perfectly intelligible at 22.05 or 16 kHz, which some transcription systems prefer. Higher rates such as 96 kHz are used in studios and add nothing for listening. Channels matter too: stereo has two, mono has one. A single-speaker lecture recorded in mono does not need stereo, and mono halves audio size. If the source has 5.1 surround sound, extraction typically downmixes to stereo for compatibility. Keep the sample rate the same as the source to avoid resampling artifacts.

Step-by-step extraction

Open the extract-audio tool and upload your video; common formats such as MP4, MOV, MKV and WebM are accepted. Choose the output format and, where offered, the bitrate. Start the extraction and wait for processing; the time depends mainly on length and file size, not video resolution. Download the audio file and open it to check that the start is not cut off and that the speakers sound natural. If you have a long file and only need part of it, use an audio cutter afterwards to keep the section you want. Rename the file clearly, with date or topic. If you need different formats, extract to the highest-quality one first and convert from there.

Common pitfalls

A video with no audio track, or with audio stored in a rare codec, may fail to produce output; check the details in a metadata viewer. Videos with several audio tracks, for example different languages, may extract only the default one. If the audio sounds out of sync when you put it back with the video, the source may have variable frame rate or a start offset; extract again from the original rather than from an edited copy. Very quiet recordings stay quiet, because extraction does not normalise volume. Do not extract to MP3 for professional editing because the lossy compression affects later processing. And do not assume the result is better than the source: extraction can preserve quality, never improve it.

Use extracted audio responsibly

Audio is protected by copyright just as video is, and separating it from the picture does not change who owns it. Extract audio from your own recordings, from meetings you took part in with the participants' agreement, from material under a licence that allows reuse, or from content where the creator has given permission. Be especially careful with commercial music and film soundtracks: using them in your own videos or sharing them can lead to takedowns and legal claims even when you only meant to keep a private copy. If you plan to transcribe a recording that contains other people's voices, tell them first. Credit sources when you quote or republish excerpts, and keep the original file as proof of provenance.

Try the free video downloader

Get formats

Related guides